Background In 1999 we discovered mutations in the gene (TNFRSF1A) encoding the 55 kDa tumor necrosis factor (TNF) alpha receptor in seven families with dominantly inherited periodic fever and inflammation. In addition to their fever and serosal inflammation, these families shared several distinguishing clinical features, including prolonged (greater than one week) attacks, a characteristic migratory erythematous rash with associated myalgia, conjunctival involvement, and a more marked therapeutic response to corticosteroids than to colchicine. This syndrome had first been described in a large Irish family, and had therefore been termed "familial Hibernian fever," but upon finding mutations not only in Irish families but in other nationalities, we suggested the more ethnically neutral name "TNF receptor-associated periodic syndrome (TRAPS)." All of the mutations we identified were missense mutations in the extracellular cysteine-rich domains of the TNF receptor, with five disrupting highly conserved disulfide bonds. For the C52F mutation, we demonstrated a defect in activation-induced ectodomain shedding of the TNF receptor, which may interfere with normal homeostasis in inflammation. During the 2000 reporting period we: (1) described eight additional mutations, including one splice mutation, in TNFRSF1A among newly referred patients. Two of these substitutions, R92Q and P46L, were found in about 1% of control chromosomes, but at even higher frequencies among periodic fever patients; (2) demonstrated receptor shedding defects for five additional mutations, but not for R92Q; (3) identified a common intragenic haplotype for apparently unrelated patients with the R92Q mutation, but not for T50M; (4) provided preliminary data that the frequency of R92Q is increased among patients in an early arthritis clinic; (5) performed linkage studies suggesting further genetic heterogeneity among the dominantly inherited periodic fevers; (6) initiated a treatment protocol to evaluate the efficacy of the anti-TNF agent etanercept in TRAPS; and (7) began constructing targeting vectors to develop TRAPS knockin mice. Results of the Last Year Mutational studies: Although we have identified a number of new cases with already known mutations, we have found only one new TNFRSF1A mutation (C43S) among the last year's referrals. This brings to 17 the total number of known mutations. All are in the extracellular domain of the receptor; one is a splicing mutation, nine are missense substitutions of cysteines, and the remainder are missense mutations at other residues. Genotype-phenotype studies performed before the discovery of C43S indicated a statistically insignificant increase in penetrance for cysteine substitutions (54/59 vs. 46/58) over other mutations. Based on our own patients and those reported in the literature, we found that the frequency of amyloidosis in TRAPS is approximately 14%, with 13 of the 14 cases occurring in patients with cysteine substitutions. In a study of 90 sporadic cases of periodic fever, complete genomic sequencing of the exons and splice junctions identified no additional mutations. A followup of our earlier studies of R92Q in early arthritis identified a total of 7 of 135 patients who bore this mutation. Etanercept study: We have now reached our accrual limit of 15 for Protocol 00-AR-0112, to examine systematically the use of etanercept in TRAPS. The trial consists of a 3 month observation period on "standard therapy" (for most patients, intermittent corticosteroids), 3 months on etanercept twice weekly (25 mg SC for adults, 0.4 mg/kg for children), 3 months on etanercept 3 times a week if there has not been a complete response to twice weekly (otherwise, a continuation of etanercept twice weekly), and a 3 month washout period. Patients are monitored for the number and severity of attacks, baseline and attack-associated levels of a number of inflammatory parameters, and the need for other medications (analgesics, corticosteroids, etc.). Of the 15 patients enrolled, 10 have the T50M mutation, 2 have H22Y, 1 has P46L, 1 has R92Q, and 1 has C33G. None have evidence of systemic amyloidosis. There have been two withdrawals: the R92Q patient withdrew for lack of efficacy, and the C33G patient was withdrawn in the first quarter because of inability to undergo periodic blood testing. Of the remaining 13, all have required a dose escalation in the third quarter. Ten have completed all four quarters, and 3 are in the washout phase. Of the 10 who have completed the study, all have had a dose-dependent decrease in their attack score and their use of other medications, which rebounded during the washout phase. Development of TRAPS knockin mice: We now have chimeric mice with the T50M mutation, and these are being tested for germline transmission. Embryonic stem cells with the C33Y and C52F mutations are currently being prepared for blastocyst injection. Development of an in vitro system to study the function of mutant forms of TNFRSF1A: We have developed expression vectors for wild-type TNFRSF1A, and for the H22Y, C30S, C33G, T50M, C52F, C88R, and R92Q mutations. Using biotinylated TNF and flow cytometry, we have found that R92Q receptors bind TNF as well as the wild type, that TNF binding to H22Y and C30S mutants is decreased but not abolished, and that TNF binding to the other 4 mutants (C33G, T50M, C52F, and C88R) is totally ablated. Experiments are in progress to examine receptor pre-ligand assembly by fluorescence energy transfer (FRET), protein size and quantity by Western analysis, signaling through the NF kappa B, jun kinase, and caspase pathways, and activation-induced receptor shedding. Conclusions and Significance The studies of the last year provide additional details in the characterization of TRAPS. Among the mutational studies, the most significant findings are the association of cysteine mutations with an increased risk of systemic amyloidosis, and the increased frequency of the R92Q mutation with early arthritis. The etanercept treatment protocol is now near completion, and it appears to confirm our earlier pilot experience that this agent reduces the frequency and severity of TRAPS attacks. The data on transfected mutant receptors strongly suggest that there may be additional mechanisms of inflammation in TRAPS other than the failure of activation-induced receptor cleavage. The observation that C33G, T50M, C52F, and C88R mutants do not bind TNF might suggest that these mutations lead to shunting of TNF signaling to p75 receptors, which lack a death domain and therefore cannot signal apoptosis. Thus, as appears to be the case for other periodic fevers, TRAPS may in part be caused by a failure of apoptosis in a subset of leukocytes involved in the early steps of the inflammatory cascade. During the next year, our objectives will be: (1) to continue mutational and genotype-phenotype studies of selected patients and families referred to the NIH Clinical Center; (2) to examine the role of TNFRSF1A sequence variants in a range of inflammatory conditions; (3) to complete our clinical trial of etanercept in TRAPS, and possibly to initiate a followup study comparing etanercept with another TNF inhibitor; (4) to begin phenotypic, biochemical, and cellular studies of T50M TRAPS knockin mice; and (5) to continue in vitro functional assays in transfected cell lines.